Xiaomi's MiMo-V2.6-Pro tops open-weights charts with quality-weighted RL over 7,000 environments
DeepLearningAI · x · 2026-10-07
Xiaomi released MiMo-V2.6-Pro-RL (1.02T-param MoE, 42B active) and Flash (309B, 15B active), topping Artificial Analysis' Intelligence Index among open weights models (46), near GPT-6 Sol at lower per-task cost. Key fix: models trained only to pass tests learned to add unrequested code and let errors pass silently — Xiaomi multiplied each test result by quality-checklist scores. They also open-sourced 7,000+ RL task environments, training code, and a Qwen3.5-9B distill. Subscriptions $6–$100/month.
More from Models
- Western labs less likely to distill from Claude but more likely from GLM 5.3 — andrew_n_carr · 2026-10-07
- With OpenAI's Luna, cloud beats open weights on price — at least GPUs heat the house — BLUECOW009 · 2026-10-07
- V4.1 Flash review: top of Flash tier but wild hallucination swings, tester wary of V4.1 Pro — teortaxesTex · 2026-10-07
- OpenAI Launches Decisions API in Public Beta, Up to 10x Faster Than GPT-6 Luna — stevenheidel · 2026-10-07
- OpenAI to watermark ChatGPT outputs by default in the EU under AI Act — Ars Technica AI · 2026-10-07
- Mistral trained ML4 on 3,800 Grace Blackwell GPUs in its European datacenter, more clusters coming — rakyll · 2026-10-07